Fwd: Concurrency control

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
17 messages Options
Reply | Threaded
Open this post in threaded view
|

Fwd: Concurrency control

Laxman Ch
Hi,

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.
- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 
- Using Hadoop version 2.6.0
- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

If needed, I can provide more details on the usecase and problem.

--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Naganarasimha Garla
Hi Laxman,
Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator" configured to DominantResourceCalculator then cpu and memory can be controlled.
Please Kindly  furhter refer to the official documentation 
http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html

But may be if say more about problem then we can suggest ideal configuration, seems like capacity configuration and splitting of the queue is not rightly done or you might refer to Fair Scheduler if you want more fairness for container allocation for different apps.

On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
Hi,

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.
- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 
- Using Hadoop version 2.6.0
- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

If needed, I can provide more details on the usecase and problem.

--
Thanks,
Laxman

Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Laxman Ch
Yes. I'm already using cgroups. Cgroups helps in controlling the resources at container level. But my requirement is more about controlling the concurrent resource usage of an application at whole cluster level.

And yes, we do configure queues properly. But, that won't help. 

For example, I have an application with a requirement of 1000 vcores. But, I wanted to control this application not to go beyond 100 vcores at any point of time in the cluster/queue. This makes that application to run longer even when my cluster is free but I will be able meet the guaranteed SLAs of other applications.

Hope this helps to understand my question.

And thanks Narasimha for quick response.

On 17 September 2015 at 16:17, Naganarasimha Garla <[hidden email]> wrote:
Hi Laxman,
Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator" configured to DominantResourceCalculator then cpu and memory can be controlled.
Please Kindly  furhter refer to the official documentation 
http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html

But may be if say more about problem then we can suggest ideal configuration, seems like capacity configuration and splitting of the queue is not rightly done or you might refer to Fair Scheduler if you want more fairness for container allocation for different apps.

On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
Hi,

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.
- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 
- Using Hadoop version 2.6.0
- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

If needed, I can provide more details on the usecase and problem.

--
Thanks,
Laxman




--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Naganarasimha Garla
Hi Laxman,
For the example you have stated may be we can do the following things : 
1. Create/modify the queue with capacity and max cap set such that its equivalent to 100 vcores. So as there is no elasticity, given application will not be using the resources beyond the capacity configured
2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that each active user would be assured with the minimum guaranteed resources . By default value is 100 implies no user limits are imposed.

Additionally we can think of "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage" which will enforce strict cpu usage for a given container if required.

+ Naga

On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
Yes. I'm already using cgroups. Cgroups helps in controlling the resources at container level. But my requirement is more about controlling the concurrent resource usage of an application at whole cluster level.

And yes, we do configure queues properly. But, that won't help. 

For example, I have an application with a requirement of 1000 vcores. But, I wanted to control this application not to go beyond 100 vcores at any point of time in the cluster/queue. This makes that application to run longer even when my cluster is free but I will be able meet the guaranteed SLAs of other applications.

Hope this helps to understand my question.

And thanks Narasimha for quick response.

On 17 September 2015 at 16:17, Naganarasimha Garla <[hidden email]> wrote:
Hi Laxman,
Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator" configured to DominantResourceCalculator then cpu and memory can be controlled.
Please Kindly  furhter refer to the official documentation 
http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html

But may be if say more about problem then we can suggest ideal configuration, seems like capacity configuration and splitting of the queue is not rightly done or you might refer to Fair Scheduler if you want more fairness for container allocation for different apps.

On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
Hi,

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.
- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 
- Using Hadoop version 2.6.0
- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

If needed, I can provide more details on the usecase and problem.

--
Thanks,
Laxman




--
Thanks,
Laxman

Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Laxman Ch
No Naga. That wont help.

I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with same user which runs in same queue (capacity=100vcores). In this scenario, if app1 triggers first occupies all the slots and runs longs then app2 will starve longer.

Let me reiterate my problem statement. I wanted "to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY"

On 17 September 2015 at 22:28, Naganarasimha Garla <[hidden email]> wrote:
Hi Laxman,
For the example you have stated may be we can do the following things : 
1. Create/modify the queue with capacity and max cap set such that its equivalent to 100 vcores. So as there is no elasticity, given application will not be using the resources beyond the capacity configured
2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that each active user would be assured with the minimum guaranteed resources . By default value is 100 implies no user limits are imposed.

Additionally we can think of "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage" which will enforce strict cpu usage for a given container if required.

+ Naga

On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
Yes. I'm already using cgroups. Cgroups helps in controlling the resources at container level. But my requirement is more about controlling the concurrent resource usage of an application at whole cluster level.

And yes, we do configure queues properly. But, that won't help. 

For example, I have an application with a requirement of 1000 vcores. But, I wanted to control this application not to go beyond 100 vcores at any point of time in the cluster/queue. This makes that application to run longer even when my cluster is free but I will be able meet the guaranteed SLAs of other applications.

Hope this helps to understand my question.

And thanks Narasimha for quick response.

On 17 September 2015 at 16:17, Naganarasimha Garla <[hidden email]> wrote:
Hi Laxman,
Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator" configured to DominantResourceCalculator then cpu and memory can be controlled.
Please Kindly  furhter refer to the official documentation 
http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html

But may be if say more about problem then we can suggest ideal configuration, seems like capacity configuration and splitting of the queue is not rightly done or you might refer to Fair Scheduler if you want more fairness for container allocation for different apps.

On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
Hi,

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.
- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 
- Using Hadoop version 2.6.0
- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

If needed, I can provide more details on the usecase and problem.

--
Thanks,
Laxman




--
Thanks,
Laxman




--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Laxman Ch
Bouncing this thread again. Any other thoughts please?

On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
No Naga. That wont help.

I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with same user which runs in same queue (capacity=100vcores). In this scenario, if app1 triggers first occupies all the slots and runs longs then app2 will starve longer.

Let me reiterate my problem statement. I wanted "to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY"

On 17 September 2015 at 22:28, Naganarasimha Garla <[hidden email]> wrote:
Hi Laxman,
For the example you have stated may be we can do the following things : 
1. Create/modify the queue with capacity and max cap set such that its equivalent to 100 vcores. So as there is no elasticity, given application will not be using the resources beyond the capacity configured
2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that each active user would be assured with the minimum guaranteed resources . By default value is 100 implies no user limits are imposed.

Additionally we can think of "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage" which will enforce strict cpu usage for a given container if required.

+ Naga

On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
Yes. I'm already using cgroups. Cgroups helps in controlling the resources at container level. But my requirement is more about controlling the concurrent resource usage of an application at whole cluster level.

And yes, we do configure queues properly. But, that won't help. 

For example, I have an application with a requirement of 1000 vcores. But, I wanted to control this application not to go beyond 100 vcores at any point of time in the cluster/queue. This makes that application to run longer even when my cluster is free but I will be able meet the guaranteed SLAs of other applications.

Hope this helps to understand my question.

And thanks Narasimha for quick response.

On 17 September 2015 at 16:17, Naganarasimha Garla <[hidden email]> wrote:
Hi Laxman,
Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator" configured to DominantResourceCalculator then cpu and memory can be controlled.
Please Kindly  furhter refer to the official documentation 
http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html

But may be if say more about problem then we can suggest ideal configuration, seems like capacity configuration and splitting of the queue is not rightly done or you might refer to Fair Scheduler if you want more fairness for container allocation for different apps.

On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
Hi,

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.
- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 
- Using Hadoop version 2.6.0
- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

If needed, I can provide more details on the usecase and problem.

--
Thanks,
Laxman




--
Thanks,
Laxman




--
Thanks,
Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

RE: Concurrency control

Rohith Sharma K S

Hi Laxman,

 

In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS , probably you should be able to achieve  your goal i.e avoiding starving of applications for resources.

 

 

org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<<a href="eclipse-javadoc:%E2%98%82=hadoop-yarn-server-resourcemanager/src%5C/main%5C/java%3Corg.apache.hadoop.yarn.server.resourcemanager.scheduler.policy%7BFairOrderingPolicy.java%E2%98%83FairOrderingPolicy%5dS">S>

An OrderingPolicy which orders SchedulableEntities for fairness (see FairScheduler FairSharePolicy), generally, processes with lesser usage are lesser. If sizedBasedWeight is set to true then an application with high demand may be prioritized ahead of an application with less usage. This is to offset the tendency to favor small apps, which could result in starvation for large apps if many small ones enter and leave the queue continuously (optional, default false)

 

 

Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463

 

Thanks & Regards

Rohith Sharma K S

 

From: Laxman Ch [mailto:[hidden email]]
Sent: 29 September 2015 13:36
To: [hidden email]
Subject: Re: Concurrency control

 

Bouncing this thread again. Any other thoughts please?

 

On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:

No Naga. That wont help.

 

I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with same user which runs in same queue (capacity=100vcores). In this scenario, if app1 triggers first occupies all the slots and runs longs then app2 will starve longer.

 

Let me reiterate my problem statement. I wanted "to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY"

 

On 17 September 2015 at 22:28, Naganarasimha Garla <[hidden email]> wrote:

Hi Laxman,

For the example you have stated may be we can do the following things : 

1. Create/modify the queue with capacity and max cap set such that its equivalent to 100 vcores. So as there is no elasticity, given application will not be using the resources beyond the capacity configured

2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that each active user would be assured with the minimum guaranteed resources . By default value is 100 implies no user limits are imposed.

 

Additionally we can think of "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage" which will enforce strict cpu usage for a given container if required.

 

+ Naga

 

On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:

Yes. I'm already using cgroups. Cgroups helps in controlling the resources at container level. But my requirement is more about controlling the concurrent resource usage of an application at whole cluster level.

 

And yes, we do configure queues properly. But, that won't help. 

 

For example, I have an application with a requirement of 1000 vcores. But, I wanted to control this application not to go beyond 100 vcores at any point of time in the cluster/queue. This makes that application to run longer even when my cluster is free but I will be able meet the guaranteed SLAs of other applications.

 

Hope this helps to understand my question.

 

And thanks Narasimha for quick response.

 

On 17 September 2015 at 16:17, Naganarasimha Garla <[hidden email]> wrote:

Hi Laxman,

Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator" configured to DominantResourceCalculator then cpu and memory can be controlled.

Please Kindly  furhter refer to the official documentation 
http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html

 

But may be if say more about problem then we can suggest ideal configuration, seems like capacity configuration and splitting of the queue is not rightly done or you might refer to Fair Scheduler if you want more fairness for container allocation for different apps.

 

On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:

Hi,

 

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

 

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.

- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 

- Using Hadoop version 2.6.0

- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

 

If needed, I can provide more details on the usecase and problem.

 

--

Thanks,
Laxman

 



 

--

Thanks,
Laxman

 



 

--

Thanks,
Laxman



 

--

Thanks,
Laxman

Reply | Threaded
Open this post in threaded view
|

RE: Concurrency control

Naganarasimha G R (Naga)
Thanks Rohith for your thoughts ,
      But i think by this configuration it might not completely solve the scenario mentioned by Laxman, As if the there is some time gap between first and and the second app then though we have fairness or priority set for apps starvation will be there.
IIUC we can think of an approach where in we can have something similar to "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can provide  the functionality like 
"yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of the queue capacity which can be configured to allow a single app to acquire more resources.  Thoughts ?

+ Naga




From: Rohith Sharma K S [[hidden email]]
Sent: Tuesday, September 29, 2015 14:07
To: [hidden email]
Subject: RE: Concurrency control

Hi Laxman,

 

In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS , probably you should be able to achieve  your goal i.e avoiding starving of applications for resources.

 

 

org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>

An OrderingPolicy which orders SchedulableEntities for fairness (see FairScheduler FairSharePolicy), generally, processes with lesser usage are lesser. If sizedBasedWeight is set to true then an application with high demand may be prioritized ahead of an application with less usage. This is to offset the tendency to favor small apps, which could result in starvation for large apps if many small ones enter and leave the queue continuously (optional, default false)

 

 

Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463

 

Thanks & Regards

Rohith Sharma K S

 

From: Laxman Ch [mailto:[hidden email]]
Sent: 29 September 2015 13:36
To: [hidden email]
Subject: Re: Concurrency control

 

Bouncing this thread again. Any other thoughts please?

 

On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:

No Naga. That wont help.

 

I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with same user which runs in same queue (capacity=100vcores). In this scenario, if app1 triggers first occupies all the slots and runs longs then app2 will starve longer.

 

Let me reiterate my problem statement. I wanted "to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY"

 

On 17 September 2015 at 22:28, Naganarasimha Garla <[hidden email]> wrote:

Hi Laxman,

For the example you have stated may be we can do the following things : 

1. Create/modify the queue with capacity and max cap set such that its equivalent to 100 vcores. So as there is no elasticity, given application will not be using the resources beyond the capacity configured

2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that each active user would be assured with the minimum guaranteed resources . By default value is 100 implies no user limits are imposed.

 

Additionally we can think of "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage" which will enforce strict cpu usage for a given container if required.

 

+ Naga

 

On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:

Yes. I'm already using cgroups. Cgroups helps in controlling the resources at container level. But my requirement is more about controlling the concurrent resource usage of an application at whole cluster level.

 

And yes, we do configure queues properly. But, that won't help. 

 

For example, I have an application with a requirement of 1000 vcores. But, I wanted to control this application not to go beyond 100 vcores at any point of time in the cluster/queue. This makes that application to run longer even when my cluster is free but I will be able meet the guaranteed SLAs of other applications.

 

Hope this helps to understand my question.

 

And thanks Narasimha for quick response.

 

On 17 September 2015 at 16:17, Naganarasimha Garla <[hidden email]> wrote:

Hi Laxman,

Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator" configured to DominantResourceCalculator then cpu and memory can be controlled.

Please Kindly  furhter refer to the official documentation 
http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html

 

But may be if say more about problem then we can suggest ideal configuration, seems like capacity configuration and splitting of the queue is not rightly done or you might refer to Fair Scheduler if you want more fairness for container allocation for different apps.

 

On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:

Hi,

 

In YARN, do we have any way to control the amount of resources (vcores, memory) used by an application SIMULTANEOUSLY.

 

- In my cluster, noticed some large and long running mr-app occupied all the slots of the queue and blocking other apps to get started.

- I'm using Capacity schedulers (using hierarchical queues and preemption disabled) 

- Using Hadoop version 2.6.0

- Did some googling around this and gone through configuration docs but I'm not able to find anything that matches my requirement.

 

If needed, I can provide more details on the usecase and problem.

 

--

Thanks,
Laxman

 



 

--

Thanks,
Laxman

 



 

--

Thanks,
Laxman



 

--

Thanks,
Laxman

Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Namikaze Minato
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:

> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Laxman Ch
Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

RE: Concurrency control

Naganarasimha G R (Naga)
Hi Laxman,
 
What i meant was,  suppose if we support and configure yarn.scheduler.capacity.<queue-path>.app-limit-factor to .25  then a single app should not take more than 25 % of resources in the queue.
This would be a more generic configuration which can be enforced by the admin, than expecting it to be configured for per app by the user.

And for Rohith's suggestion of FairOrdering policy , I think it should solve the problem if the App which is submitted first is not already hogged all the queue's resources.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:03
To: [hidden email]
Subject: Re: Concurrency control

Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Laxman Ch
IMO, its better to have a application level configuration than to have a scheduler/queue level configuration.
Having a queue level configuration will restrict every single application that runs in that queue.
But, we may want to configure these limits for only some set of jobs and also for every application these limits can be different.

FairOrdering policy thing, order of jobs can't be enforced as these are adhoc jobs and scheduled/owned independently by different teams.

On 29 September 2015 at 16:43, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,
 
What i meant was,  suppose if we support and configure yarn.scheduler.capacity.<queue-path>.app-limit-factor to .25  then a single app should not take more than 25 % of resources in the queue.
This would be a more generic configuration which can be enforced by the admin, than expecting it to be configured for per app by the user.

And for Rohith's suggestion of FairOrdering policy , I think it should solve the problem if the App which is submitted first is not already hogged all the queue's resources.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:03

To: [hidden email]
Subject: Re: Concurrency control

Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

RE: Concurrency control

Naganarasimha G R (Naga)
Hi Laxman,

Ideally i understand it would be better its available @ application level, but  its like each user is expected to ensure that he gives the right configuration which is within the limits of max capacity.
And what if user submits some app (kind of a query execution app) with out this setting or he doesn't know how much it should take ? In general, users specifying resources for containers itself is a difficult task.
And it might not be right to expect that the admin will do it for each application in the queue either.  Basically governing will be difficult if its not enforced from queue/scheduler side.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:52
To: [hidden email]
Subject: Re: Concurrency control

IMO, its better to have a application level configuration than to have a scheduler/queue level configuration.
Having a queue level configuration will restrict every single application that runs in that queue.
But, we may want to configure these limits for only some set of jobs and also for every application these limits can be different.

FairOrdering policy thing, order of jobs can't be enforced as these are adhoc jobs and scheduled/owned independently by different teams.

On 29 September 2015 at 16:43, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,
 
What i meant was,  suppose if we support and configure yarn.scheduler.capacity.<queue-path>.app-limit-factor to .25  then a single app should not take more than 25 % of resources in the queue.
This would be a more generic configuration which can be enforced by the admin, than expecting it to be configured for per app by the user.

And for Rohith's suggestion of FairOrdering policy , I think it should solve the problem if the App which is submitted first is not already hogged all the queue's resources.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:03

To: [hidden email]
Subject: Re: Concurrency control

Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Laxman Ch
Hi Naga,

Like most of the app-level configurations, admin can configure the defaults which user may want override at application level.

If this is at queue-level then all applications in a queue will have the same limits. But all our applications in a queue may not have same SLA and we may need to restrict them differently. This requires again splitting queues further which I feel is more overhead.


On 30 September 2015 at 09:00, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,

Ideally i understand it would be better its available @ application level, but  its like each user is expected to ensure that he gives the right configuration which is within the limits of max capacity.
And what if user submits some app (kind of a query execution app) with out this setting or he doesn't know how much it should take ? In general, users specifying resources for containers itself is a difficult task.
And it might not be right to expect that the admin will do it for each application in the queue either.  Basically governing will be difficult if its not enforced from queue/scheduler side.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:52

To: [hidden email]
Subject: Re: Concurrency control

IMO, its better to have a application level configuration than to have a scheduler/queue level configuration.
Having a queue level configuration will restrict every single application that runs in that queue.
But, we may want to configure these limits for only some set of jobs and also for every application these limits can be different.

FairOrdering policy thing, order of jobs can't be enforced as these are adhoc jobs and scheduled/owned independently by different teams.

On 29 September 2015 at 16:43, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,
 
What i meant was,  suppose if we support and configure yarn.scheduler.capacity.<queue-path>.app-limit-factor to .25  then a single app should not take more than 25 % of resources in the queue.
This would be a more generic configuration which can be enforced by the admin, than expecting it to be configured for per app by the user.

And for Rohith's suggestion of FairOrdering policy , I think it should solve the problem if the App which is submitted first is not already hogged all the queue's resources.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:03

To: [hidden email]
Subject: Re: Concurrency control

Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Harsh J-2
If all your Apps are MR, then what you are looking for is MAPREDUCE-5583 (it can be set per-job).

On Thu, Oct 1, 2015 at 3:03 PM Laxman Ch <[hidden email]> wrote:
Hi Naga,

Like most of the app-level configurations, admin can configure the defaults which user may want override at application level.

If this is at queue-level then all applications in a queue will have the same limits. But all our applications in a queue may not have same SLA and we may need to restrict them differently. This requires again splitting queues further which I feel is more overhead.


On 30 September 2015 at 09:00, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,

Ideally i understand it would be better its available @ application level, but  its like each user is expected to ensure that he gives the right configuration which is within the limits of max capacity.
And what if user submits some app (kind of a query execution app) with out this setting or he doesn't know how much it should take ? In general, users specifying resources for containers itself is a difficult task.
And it might not be right to expect that the admin will do it for each application in the queue either.  Basically governing will be difficult if its not enforced from queue/scheduler side.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:52

To: [hidden email]
Subject: Re: Concurrency control

IMO, its better to have a application level configuration than to have a scheduler/queue level configuration.
Having a queue level configuration will restrict every single application that runs in that queue.
But, we may want to configure these limits for only some set of jobs and also for every application these limits can be different.

FairOrdering policy thing, order of jobs can't be enforced as these are adhoc jobs and scheduled/owned independently by different teams.

On 29 September 2015 at 16:43, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,
 
What i meant was,  suppose if we support and configure yarn.scheduler.capacity.<queue-path>.app-limit-factor to .25  then a single app should not take more than 25 % of resources in the queue.
This would be a more generic configuration which can be enforced by the admin, than expecting it to be configured for per app by the user.

And for Rohith's suggestion of FairOrdering policy , I think it should solve the problem if the App which is submitted first is not already hogged all the queue's resources.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:03

To: [hidden email]
Subject: Re: Concurrency control

Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Laxman Ch
Thanks and Perfect Harsh. Exactly what I am looking for. Most of our applications are MR.
So, this should be sufficient for us. These configurations, I will give a try and post my findings again here. Thanks again.

Thanks Naga, Rohit & Lloyd for your suggestions and discussion.

On 2 October 2015 at 07:37, Harsh J <[hidden email]> wrote:
If all your Apps are MR, then what you are looking for is MAPREDUCE-5583 (it can be set per-job).

On Thu, Oct 1, 2015 at 3:03 PM Laxman Ch <[hidden email]> wrote:
Hi Naga,

Like most of the app-level configurations, admin can configure the defaults which user may want override at application level.

If this is at queue-level then all applications in a queue will have the same limits. But all our applications in a queue may not have same SLA and we may need to restrict them differently. This requires again splitting queues further which I feel is more overhead.


On 30 September 2015 at 09:00, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,

Ideally i understand it would be better its available @ application level, but  its like each user is expected to ensure that he gives the right configuration which is within the limits of max capacity.
And what if user submits some app (kind of a query execution app) with out this setting or he doesn't know how much it should take ? In general, users specifying resources for containers itself is a difficult task.
And it might not be right to expect that the admin will do it for each application in the queue either.  Basically governing will be difficult if its not enforced from queue/scheduler side.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:52

To: [hidden email]
Subject: Re: Concurrency control

IMO, its better to have a application level configuration than to have a scheduler/queue level configuration.
Having a queue level configuration will restrict every single application that runs in that queue.
But, we may want to configure these limits for only some set of jobs and also for every application these limits can be different.

FairOrdering policy thing, order of jobs can't be enforced as these are adhoc jobs and scheduled/owned independently by different teams.

On 29 September 2015 at 16:43, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,
 
What i meant was,  suppose if we support and configure yarn.scheduler.capacity.<queue-path>.app-limit-factor to .25  then a single app should not take more than 25 % of resources in the queue.
This would be a more generic configuration which can be enforced by the admin, than expecting it to be configured for per app by the user.

And for Rohith's suggestion of FairOrdering policy , I think it should solve the problem if the App which is submitted first is not already hogged all the queue's resources.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:03

To: [hidden email]
Subject: Re: Concurrency control

Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman
Reply | Threaded
Open this post in threaded view
|

Re: Concurrency control

Naganarasimha Garla
Thanks Harsh J for the useful sharing of the info, but can we think of some way to support this scenario from YARN side ? 
like the queue configuration i mentioned or in the way Laxman mentioned (app specific override)  ? 

On Fri, Oct 2, 2015 at 10:26 PM, Laxman Ch <[hidden email]> wrote:
Thanks and Perfect Harsh. Exactly what I am looking for. Most of our applications are MR.
So, this should be sufficient for us. These configurations, I will give a try and post my findings again here. Thanks again.

Thanks Naga, Rohit & Lloyd for your suggestions and discussion.

On 2 October 2015 at 07:37, Harsh J <[hidden email]> wrote:
If all your Apps are MR, then what you are looking for is MAPREDUCE-5583 (it can be set per-job).

On Thu, Oct 1, 2015 at 3:03 PM Laxman Ch <[hidden email]> wrote:
Hi Naga,

Like most of the app-level configurations, admin can configure the defaults which user may want override at application level.

If this is at queue-level then all applications in a queue will have the same limits. But all our applications in a queue may not have same SLA and we may need to restrict them differently. This requires again splitting queues further which I feel is more overhead.


On 30 September 2015 at 09:00, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,

Ideally i understand it would be better its available @ application level, but  its like each user is expected to ensure that he gives the right configuration which is within the limits of max capacity.
And what if user submits some app (kind of a query execution app) with out this setting or he doesn't know how much it should take ? In general, users specifying resources for containers itself is a difficult task.
And it might not be right to expect that the admin will do it for each application in the queue either.  Basically governing will be difficult if its not enforced from queue/scheduler side.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:52

To: [hidden email]
Subject: Re: Concurrency control

IMO, its better to have a application level configuration than to have a scheduler/queue level configuration.
Having a queue level configuration will restrict every single application that runs in that queue.
But, we may want to configure these limits for only some set of jobs and also for every application these limits can be different.

FairOrdering policy thing, order of jobs can't be enforced as these are adhoc jobs and scheduled/owned independently by different teams.

On 29 September 2015 at 16:43, Naganarasimha G R (Naga) <[hidden email]> wrote:
Hi Laxman,
 
What i meant was,  suppose if we support and configure yarn.scheduler.capacity.<queue-path>.app-limit-factor to .25  then a single app should not take more than 25 % of resources in the queue.
This would be a more generic configuration which can be enforced by the admin, than expecting it to be configured for per app by the user.

And for Rohith's suggestion of FairOrdering policy , I think it should solve the problem if the App which is submitted first is not already hogged all the queue's resources.

+ Naga


From: Laxman Ch [[hidden email]]
Sent: Tuesday, September 29, 2015 16:03

To: [hidden email]
Subject: Re: Concurrency control

Thanks Rohit, Naga and Lloyd for the responses.

I think Laxman should also tell us more about which application type he is running.

We run mr jobs mostly with default core/memory allocation (1 vcore, 1.5GB).
Our problem is more about controlling the resources used simultaneously by all running containers at any given point of time per application.

Example: 
1. App1 and App2 are two MR apps.
2. App1 and App2 belong to same queue (capacity: 100 vcores, 150 GB).
3. Each App1 task takes 8 hrs for completion
4. Each App2 task takes 5 mins for completion
5. App1 triggered at time "t1" and using all the slots of queue.
6. App2 triggered at time "t2" (where t2 > t1) and waits longer fot App1 tasks to release the resources.
7. We can't have preemption enabled as we don't want to lose the work completed so far by App1.
8. We can't have separate queues for App1 and App2 as we have lots of jobs like this and it will explode the number of queues.
9. We use CapacityScheduler.

In this scenario, if I can control App1 concurrent usage limits to 50vcores and 75GB, then App1 may take longer time to finish but there won't be any starvation for App2 (and other jobs running in same queue)

@Rohit, FairOrdering policy may not solve this starvation problem.

@Naga, I couldn't think through the expected behavior of "yarn.scheduler.capacity.<queue-path>.app-limit-factor"
I will revert on this.

On 29 September 2015 at 14:57, Namikaze Minato <[hidden email]> wrote:
I think Laxman should also tell us more about which application type
he is running. The normal use cas of MAPREDUCE should be working as
intended, but if he has for example one MAP using 100 vcores, then the
second map will have to wait until the app completes. Same would
happen if the applications running were spark, as spark does not free
what is allocated to it.

Regards,
LLoyd

On 29 September 2015 at 11:22, Naganarasimha G R (Naga)
<[hidden email]> wrote:
> Thanks Rohith for your thoughts ,
>       But i think by this configuration it might not completely solve the
> scenario mentioned by Laxman, As if the there is some time gap between first
> and and the second app then though we have fairness or priority set for apps
> starvation will be there.
> IIUC we can think of an approach where in we can have something similar to
> "yarn.scheduler.capacity.<queue-path>.user-limit-factor"  where in it can
> provide  the functionality like
> "yarn.scheduler.capacity.<queue-path>.app-limit-factor" : The multiple of
> the queue capacity which can be configured to allow a single app to acquire
> more resources.  Thoughts ?
>
> + Naga
>
>
>
> ________________________________
> From: Rohith Sharma K S [[hidden email]]
> Sent: Tuesday, September 29, 2015 14:07
> To: [hidden email]
> Subject: RE: Concurrency control
>
> Hi Laxman,
>
>
>
> In Hadoop-2.8(Not released  yet),  CapacityScheduler provides configuration
> for configuring ordering policy.  By configuring FAIR_ORDERING_POLICY in CS
> , probably you should be able to achieve  your goal i.e avoiding starving of
> applications for resources.
>
>
>
>
>
> org.apache.hadoop.yarn.server.resourcemanager.scheduler.policy.FairOrderingPolicy<S>
>
> An OrderingPolicy which orders SchedulableEntities for fairness (see
> FairScheduler FairSharePolicy), generally, processes with lesser usage are
> lesser. If sizedBasedWeight is set to true then an application with high
> demand may be prioritized ahead of an application with less usage. This is
> to offset the tendency to favor small apps, which could result in starvation
> for large apps if many small ones enter and leave the queue continuously
> (optional, default false)
>
>
>
>
>
> Community Issue Id :  https://issues.apache.org/jira/browse/YARN-3463
>
>
>
> Thanks & Regards
>
> Rohith Sharma K S
>
>
>
> From: Laxman Ch [mailto:[hidden email]]
> Sent: 29 September 2015 13:36
> To: [hidden email]
> Subject: Re: Concurrency control
>
>
>
> Bouncing this thread again. Any other thoughts please?
>
>
>
> On 17 September 2015 at 23:21, Laxman Ch <[hidden email]> wrote:
>
> No Naga. That wont help.
>
>
>
> I am running two applications (app1 - 100 vcores, app2 - 100 vcores) with
> same user which runs in same queue (capacity=100vcores). In this scenario,
> if app1 triggers first occupies all the slots and runs longs then app2 will
> starve longer.
>
>
>
> Let me reiterate my problem statement. I wanted "to control the amount of
> resources (vcores, memory) used by an application SIMULTANEOUSLY"
>
>
>
> On 17 September 2015 at 22:28, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> For the example you have stated may be we can do the following things :
>
> 1. Create/modify the queue with capacity and max cap set such that its
> equivalent to 100 vcores. So as there is no elasticity, given application
> will not be using the resources beyond the capacity configured
>
> 2. yarn.scheduler.capacity.<queue-path>.minimum-user-limit-percent   so that
> each active user would be assured with the minimum guaranteed resources . By
> default value is 100 implies no user limits are imposed.
>
>
>
> Additionally we can think of
> "yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage"
> which will enforce strict cpu usage for a given container if required.
>
>
>
> + Naga
>
>
>
> On Thu, Sep 17, 2015 at 4:42 PM, Laxman Ch <[hidden email]> wrote:
>
> Yes. I'm already using cgroups. Cgroups helps in controlling the resources
> at container level. But my requirement is more about controlling the
> concurrent resource usage of an application at whole cluster level.
>
>
>
> And yes, we do configure queues properly. But, that won't help.
>
>
>
> For example, I have an application with a requirement of 1000 vcores. But, I
> wanted to control this application not to go beyond 100 vcores at any point
> of time in the cluster/queue. This makes that application to run longer even
> when my cluster is free but I will be able meet the guaranteed SLAs of other
> applications.
>
>
>
> Hope this helps to understand my question.
>
>
>
> And thanks Narasimha for quick response.
>
>
>
> On 17 September 2015 at 16:17, Naganarasimha Garla
> <[hidden email]> wrote:
>
> Hi Laxman,
>
> Yes if cgroups are enabled and "yarn.scheduler.capacity.resource-calculator"
> configured to DominantResourceCalculator then cpu and memory can be
> controlled.
>
> Please Kindly  furhter refer to the official documentation
> http://hadoop.apache.org/docs/r1.2.1/capacity_scheduler.html
>
>
>
> But may be if say more about problem then we can suggest ideal
> configuration, seems like capacity configuration and splitting of the queue
> is not rightly done or you might refer to Fair Scheduler if you want more
> fairness for container allocation for different apps.
>
>
>
> On Thu, Sep 17, 2015 at 4:10 PM, Laxman Ch <[hidden email]> wrote:
>
> Hi,
>
>
>
> In YARN, do we have any way to control the amount of resources (vcores,
> memory) used by an application SIMULTANEOUSLY.
>
>
>
> - In my cluster, noticed some large and long running mr-app occupied all the
> slots of the queue and blocking other apps to get started.
>
> - I'm using Capacity schedulers (using hierarchical queues and preemption
> disabled)
>
> - Using Hadoop version 2.6.0
>
> - Did some googling around this and gone through configuration docs but I'm
> not able to find anything that matches my requirement.
>
>
>
> If needed, I can provide more details on the usecase and problem.
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
>
>
> --
>
> Thanks,
> Laxman
>
>
>
>
>
> --
>
> Thanks,
> Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman



--
Thanks,
Laxman